by desktop » 30 Jul 2011 03:39
Sorry I haven't been around for 2 days spent in hospital for routine metric testing. I seem to be healthy "as an ox" to use a trite old cliche.
I fail to see how the AES test subjects in the 70s/80s were getting such high guessing scores on whether lo pass filters with various cut-off frequencies like 30KHz, 35KHz, 50KHz etc. were inserted into a signal path or not, in blind listening tests, were tuning into IM products or other non-high-frequency sensing, because then all the engineers could be doing it and getting great scores on their ability to choose whether the filters were there or not.
Please read the simple description by Tannoy on their website, that described the leading edge for a percussive 30Hz bass note (like a major drum whack) as requiring a loudspeaker to be capable of 30KHz performance, which only a few loudspeakers can do. Then see how Tannoy of all the "old fart" manufacturers, did a complete turn-around from total nay-sayers to true believers once they did their own research into ultra-high frequency reproduction.
I suggest that to produce a group of facts that we can discuss, a critical mass of people should actually listen to a speaker with measured performance out to 40KHz. If we as a group of, either at least 17 people, or if possible 300-ish people, listen, then their listening impressions would be facts. As in, "I listened to this product, I could/couldn't hear something, here is my educated opinion."
I find it hard to believe the number of nay-sayers there are who put down an important manufacturer's product (based on the facts of Tannoy's own research), while never having heard this product. For some reason the whole concept that extended high frequency loudspeakers, and listeners who can sense these frequencies are somehow scary, or the concept ruins so many of the nay-sayers preconceptions that they hold so dear, cause those nay-sayers to say anything to discourage open-minded listeners from listening to these products. If possible please use facts based on tests that people have done using a wideband, loudspeakers system, source material, and amplifier throughput systems, on listeners who are open minded enough to consider that they COULD sense ultra-high frequencies.
The thing I find most worrying is that given the various possibilities to explain why these 1970s/80s AES bandwidth listening non-guesses were right so often, is that few (but me it seems) are willing to consider that when 30 AES test subject can all get 90+% scores on filters inserted or not at 45KHz, one of those possibilities is that the engineers taking the tests really could sense the difference between a lo pass filter inserted at 45KHz and no filter at all.
The last paper used as an example of how no one on earth can really hear frequencies about 25KHz was done in the 1950s, when there were no transducers of quality that could reproduce complex signals (not tones) above 25KHz with distortion under 20%. So what were these test paper doctors using as their reproducersto play back their test signals?
While we all know there are snake oil salesmen with pyramids you put on amplifiers, or magic mats for TTs, the number of qualified engineers who support the Sony/Philips/Tannoy perspective numbers in the thousands at this point. Have any of the nay-sayers asked any of these engineers if they can hear a difference if high frequencies are extend to 40KHz.
Sounds like the answer from the nay-says is
"Nope, we wouldn't ask someone who actually uses these products because they might know something we don't know, and if our preconceptions were to crumble then we'd look like fools. And we certainly wouldn't listen to these products ourselves because then if it turns out that ultra-high frequency production can enhance the realism of a program material, we'd know we were wrong all along."
So the best way for the nay-sayers to attack this possibility (that ultra high frequency energy reproduction in all levels of the audio chain from recording the instruments, to reproducing it in a listening environment, does increase the reality of the program materials), is to avoid learning anything about the products that perform in the ultra-high frequency range, and avoid all input from the people who actually use these products (no matter how qualifies the users are, as well as discouraging others from listening to these wideband loudspeakers).
But I am used to this type of resistance to any new possibility, challenging old misconceptions. The test data from the 50s may no longer be valid, because the equipment in audio is so much better now. The better way to test this idea of whether ultra-high frequencies can enhance the realism of any program material is to do what Sony/Philips/Tannoy are doing. They have many customers who are trying out their prototype products, and in addition, various studio owners are willing to let others in their local AES community also listen to this "new thing". Luckily most of these studio engineers are not nay-sayers. These engineers want to find out for themselves if these kind of extended high frequency capable SYSTEMS, results in more realism and better program material production.
I find that musicians like Katz and Becker were never nay-sayers, always trying to learn more about sound and studio production, so that Steely Dan albums could keep improving. If you listen to the series of Steely Dan albums, and you will see that the productions kept getting better all the time. This is why ABC Records/Steely Dan helped in producing for ESS, a group of Aja master recordings to use in the test of 17,000 students around America.
First those guys went out on their own and listened carefull to the ESS STudio Monitor with high frequency response not only extended to past 30KHz, but creating a dispersion pattern of up/down 25 degree, and Left/Right about 50 degrees at 25KHz. This was a rather remarkable performance for a product developed in 1973 w/the ESS AMT-1. After they listened to this product on their own, and felt it was representative of what could be possible at ABC Studios on Beverly Blvd. They agreed to a second step in the process. After the listening tapes using ESS Studio 12 monitors were made at ABC Studios, and approved for the nationwide listening test, we moved to the JVC Cutting Lab to see how the ESS monitors sound with vinyl.
I still have the test pressings of Gaucho that were used for this listening test (pressed only on one side). The ESS monitor was a huge success because it allowed everyone involved to hear how well a musical production could sound if the monitors produced ultra-high frequency energy out to beyond 25KHz. A few minor changes were made on Gaucho at that time on the final disc mastering tapes. Of course the ESS monitor was not durable enough for studio use, but only because the ESS sound reinforcement tweeters were not used. Even so I would take Katz and Becker's opinion on extended high frequency response as equal to my own experience and superior to anyone who has never even tried to listen to wideband program material, on a loudspeaker capable of wideband frequency reproduction, because some people reserve their own opinions to sound they have actual experience with.
But there is hope for the nay-sayers with no experience listening to the products they are putting down. Tannoy of all manufacturers were of the opinion in the 70s and 80s that there would be no purpose in making a wideband loudspeaker system. They made a nice concentric loudspeaker with response to about 17/18 KHz and they couldn't imagine that wideband sound could make a difference to listeners.
But now Tannoy's own study has proven their previous opinions to be dead wrong. With program material that has extended high frequency, an amplification system that is not purposely band limited (like the various Quad power amps starting with the 303), and a loudspeaker that also has extended high frequency reproduction ability, the sound produced with wideband program material is much better on wideband loudspeakers than speakers with 20/22KHz usable frequency cut-offs.
I prefer discussions with people who have personal experience with the sound products being discussed. I could argue with someone about the Martin Logan Statement series of loudspeakers, that I have heard perhaps for a total of 1 hour, but it would all be so pointless. The room was what it was, the program material was what it was, the amp was very nice (a Conrad Johnson 200w monoblock), but all these factors were rather unfamiliar to me. SO what kind of judgment can I really make. Of course when someone who has never heard the Statements AT ALL, EVER, has numerous arguments about how that speaker could not possibly be worth what the price was, I wonder what this person could possibly know about what he is attempting to denigrate. IMO such opinions aren't worth anything.
But there may be a few listeners who do go to listen to the Tannoy Ellipse 8 or 10. If they can stand the heat, they may even comment on what they heard (or sensed) in the way of ultra-high frequency content. It is worth something (though not as much as actual experience), if someone chats with a studio engineer who actually uses, or after a long listening trial, turned down, the Tannoy Ellipse 8 or 10. In this case we are talking about second hand opinions, but that's better than no basis in product listening when forming an opinion.
People today are pretty cynical about product evaluations when they pay big money to buy something my detractors say has meaningless high frequency output. The internet provides them with a venue to voice their opinions. I haven't found people with good hearing saying that the Tannoy Ellipse 8 or 10 tweeters are rubbish, and not worth having. On the contrary, continuing sales keeps giving the opposite impression. I'm willing to discuss with experienced listeners, the benefits and down-side factors in the Ellipse 8s or 10s they have had long listening sessions using. But short of that I find that the use of old obsolete data from the 50s or other conjecture NOT based on actual listening to be pretty pointless.
When some researcher does the next "modern" tests to see whether wideband loudspeakers produce greater realism playing back recorded material (given good hearing for the listener, a loudspeaker capable of low distortion extended frequency output capability, and an amplified thoughput system that is not band limited), I feel confident that my opinions about ultra-high frequency output ability in loudspeakers will be born out. I've seen the results in action beginning with the CV loudspeakers with thin film tweeters, the ESS loudspeaker which was perhaps the best HF loudspeaker, and many others.
My suggestion to anyone who wants to discuss loudspeakers with super-extended high frequency response INTELLIGENTLY, is to listen to such products. Otherwise you don't know what you are talking about.
I have lived in the audio business when the elliptical stylus was "the best", but then was challenged by the shibata, hyper-elliptical and line contact styli, etc. etc. The difference between these last 3 is primarily in the quality of lower distortion, ultra-high frequency response. With gem cantilevers (sapphire, diamond & ruby) the extended highs were improved again by +20KHz resonances for cartridges like the Ortofon Kontrapunkt B or the Dynavector 17D3. Now that we have cartridges that can reproduce ultra-high frequencies with low distortion, we should have loudspeakers/amps that can reproduce those same ultra-high frequencies, but lower distortion if possible.
